|Publication number||US7286984 B1|
|Application number||US 10/158,082|
|Publication date||Oct 23, 2007|
|Filing date||May 31, 2002|
|Priority date||Nov 5, 1999|
|Also published as||US7440897, US7620548, US8010361, US8200491, US20080046243, US20080177544, US20080215328, US20080288244, US20110313769|
|Publication number||10158082, 158082, US 7286984 B1, US 7286984B1, US-B1-7286984, US7286984 B1, US7286984B1|
|Inventors||Allen Louis Gorin, Dijana Petrovska-Delacretaz, Giuseppe Riccardi, Jeremy Huntley Wright|
|Original Assignee||At&T Corp.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (48), Non-Patent Citations (22), Referenced by (24), Classifications (5), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This non-provisional application claims the benefit of U.S. Provisional Patent Application No. 60/322,447, filed Sep. 17, 2001, which is incorporated herein by reference in its entirety. This application is also a continuation-in-part of U.S. patent application Ser. Nos. 09/690,721 and 09/690,903 now U.S. Pat. No. 6,681,206 both filed Oct. 18, 2000, which claim priority from U.S. Provisional Application No. 60/163,838, filed Nov. 5, 1999. U.S. patent application Ser. Nos. 09/690,721, 09/690,903 and U.S. Provisional Application No. 60/163,838 are incorporated herein by reference in their entireties.
The invention relates to automated systems for communication recognition and understanding.
Conventional methods for constructing and training statistical models for recognition and understanding involve collecting and annotating large speech corpora for a task. This speech is manually transcribed and each utterance is then semantically labeled. The resultant database is exploited to train stochastic language models for recognition and understanding. These models are further adapted for different dialog states. Examples of such methods are shown in U.S. Pat. Nos. 5,675,707, 5,860,063, 6,044,337, 6,192,110, and 6,173,261, each of which is incorporated by reference herein in its entirety.
This transcription and labeling process is a major bottleneck in new application development and refinement of existing ones. For incremental training of a deployed automated dialog system, current technology would potentially require transcribing millions of transactions. This process is both time-consuming and prohibitively expensive.
The invention concerns a method and system for detecting morphemes in a user's communication. The method may include recognizing a lattice of phone strings from the user's input communication, the lattice representing a distribution over the phone strings, and detecting morphemes in the user's input communication using the lattice.
The morphemes may be acoustic and/or non-acoustic. The morphemes may represent any unit or sub-unit of communication including phones, diphones, phone-phrases, syllables, grammars, words, gestures, tablet strokes, body movements, mouse clicks, etc. The training speech may be verbal, non-verbal, a combination of verbal and non-verbal, or multimodal.
The invention is described in detail with reference to the following drawings wherein like numerals reference like elements, and wherein:
This invention concerns a dialog system that automatically learns from speech without transcriptions. Semantic labels can be extracted automatically from either experiments or from autonomous dialogs. In particular, a task-independent phone-recognizer is taught how to ‘learn to understand’ from a database of untranscribed (or transcribed) speech plus semantic labels.
Baseline approaches to the teaching of a speech recognition systems to understand are found in U.S. Pat. Nos. 5,675,707, 5,860,063, 6,044,337, 6,192,110 and 6,173,261, which are incorporated herein by reference in their entireties.
The earliest work demonstrated automatic acquisition of ‘words’ and ‘grammar’ from collapsed text. That work did not address, however, the issues arising from non-perfect recognition of speech. The next step was to show how to acquire lexical units from speech alone without transcription and exploit them for spoken language understanding (SLU). That experiment, however, was constrained to speech comprising isolated word sequences and used matching techniques to decide if an observation was a new ‘word’ or variation of a known ‘word’.
All of the above efforts involve learning from speech alone. While one can learn much about a spoken language by merely listening to it, the process can progress further and faster by exploiting semantics. This has been demonstrated in both the engineering domain and in analyses of children's language acquisition. Thus, this invention goes beyond the past efforts by exploiting speech plus meaning using morphemes, both acoustic and non-acoustic, in order to teach a machine to learn to understand.
While the morphemes may be non-acoustic (i.e., made up of non-verbal sub-morphemes such as tablet strokes, gestures, body movements, etc.), for ease of discussion, the systems and methods illustrated in the drawings and discussed in the below concern only acoustic morphemes. Consequently, the invention should not be limited to just acoustic morphemes and should encompass the utilization of any sub-units of any known or future method of communication for the purposes of recognition and understanding.
Furthermore, while the terms “speech”, “phrase” and “utterance”, used throughout the description below, may connote only spoken language, it is important to note in the context of this invention, “speech”, “phrase” and “utterance” may include verbal and/or non-verbal sub-units (or sub-morphemes). Therefore, “speech”, “phrase” and “utterance” may comprise non-verbal sub-units, verbal sub-units or a combination of verbal and non-verbal sub-units within the sprit and scope of this invention.
The morpheme generation subsystem 110 includes a morpheme generator 130 and a morpheme database 140. The morpheme generator 130 generates morphemes from a corpus of untranscribed training speech (the invention may also operate with training speech that is transcribed, of course). The generated morphemes are stored in the morpheme database 140 for use by the morpheme detector 160. The morpheme database 140 contains a large number of verbal and non-verbal speech fragments or sub-morphemes (illustrated as phone-phrases for ease of discussion), each of which is related to one or more of a predetermined set of task objectives. Each of the morphemes may be labeled with its associated task objective. The operation of the morpheme generator 130 will described in greater detail with respect to
The input speech classification subsystem 120 includes an input speech recognizer 150, a morpheme detector 160 and a task classification processor 170. The input speech recognizer 150 receives a user's task objective request in the form of verbal and/or non-verbal speech. The input speech recognizer 150 may perform the function of recognizing, or spotting the existence of one or more phones, sub-units, acoustic morphemes, etc. in the user's input speech by any algorithm known to one of ordinary skill in the art. However, the input speech recognizer 150 forms a lattice structure to represent a distribution of recognized phone sequences, such as a probability distribution. The input speech recognizer 150 may extract the n-best phone strings that may be extracted from the lattice, either by themselves or along with their confidence scores. Lattice representations are well known those skilled in the art and are further described in detail below.
While the method of morpheme detection using lattices is shown in the figures as being associated with a task classification system, this is purely exemplary. The method of morpheme detection using lattices may be applied to a wide variety of automated communication systems, including customer care systems, and should not be limited to a task classification system.
The morpheme detector 160 then detects the acoustic and/or non-acoustic morphemes present in the lattice that represents the user's input request. The morphemes generated by the morpheme generation subsystem 110 are provided as an input to the morpheme detector 160.
The output of morpheme detector 160 includes the detected morphemes appearing in the user's task objective request that is then provided to the task classification processor 170. The task classification processor 170 may apply a confidence function, based on the probabilistic relation between the recognized morphemes and selected task objectives, and makes a decision either to implement a particular task objective, or makes a determination that no decision is likely in which case the user may be defaulted to a human or automated system for assistance.
An exemplary process of the invention will now be described with reference to
In an embodiment for recognizing non-acoustic morphemes, the ASR phone recognizer 210 may be replaced in the figure by a sub-morpheme recognizer. The sub-morpheme recognizer would operate similar to the ASR phone recognizer, but it would receive raw non-acoustic or a mixture of acoustic and non-acoustic training data from a database. However, for each of discussion, use of only acoustic morphemes will be described below.
In addition, while the drawings illustrate the use of phones, this is purely exemplary. Any sub-portion of verbal and/or non-verbal speech may by recognized and detected within the spirit and scope of the invention.
A training set of thousands of spoken utterances with corresponding call-labels is used, followed by using a separate test set in the range of 1000 utterances. They are designated HHS-train and HHS-test, respectively. HHS denotes human/human speech-only.
The ASR phone recognizer 210 that is applied to the training speech is task-independent. In particular, a phonotactic language model was trained on the switchboard corpus using a Variable-Length N-gram Stochastic Automaton. This corpus is unrelated to the HMIHY task, except in that they both comprise fluent English speech. Off-the-shelf telephony acoustic models may be used. Applying the ASR phone recognizer 210 to the HMIHY test speech data yields a phone accuracy of 43%. The training and test sets so generated are denoted by ASR-phone-train and ASR-phone-test respectively.
For a baseline comparison, a ‘noiseless’ phonetic transcription was generated from the orthographic transcriptions, by replacing each word by its most likely dictionary pronunciation and deleting word-delimiters. E.g. “collect call” is converted to “K ax I eh K T K ao I” (see
The number of recognized phones per utterance is distributed as shown in
For each utterance, the length of the recognized phone sequence is compared with the length of the phonetic transcription. These values are scatter-plotted in
In step 3200, the salient phone-phrase generator 220 selects candidate phone-phrases from the raw training speech corpus. While the system and method of the invention is illustrated and described using the term phone-phrases, it is again important to note that phone-phrases are actually sub-morphemes that may be acoustic or non-acoustic (i.e., made up of non-verbal sub-morphemes such as tablet strokes, gestures, body movements, etc.). However, as discussed above, for ease of discussion, the systems and methods illustrated in the drawings and discussed in the below concern only phone-phrases. Consequently, the invention should not be limited to using just phone-phrases and should encompass the utilization of any sub-units of any known or future method of communication for the purposes of recognition and understanding.
Once the candidate phone-phrases have been selected, in step 3300, the salient phone-phrase generator 220 selects a subset of the candidate phone-phrases. Thus, new units are acquired by the above process of searching the space of observed phone-sequences and selecting a subset according to their utility for recognition and understanding. The resultant subset selected is denoted as salient phone-phrases. Examples of salient phone-phrases for the word “collect” are shown in
The salient phone-phrase generator 220 may perform the selection of salient phone-phrases by first using a simplified measure of the candidate phone-phrase's salience for the task as the maximum of the a posteriori distribution,
where C varies over the 15 call-types in the HMIHY task. The salient phone-phrases are then selected by applying a threshold on Pmax and by using a multinomial statistical significance test. This significance test excludes low-frequency phrases for which a fortunate conjunction of events can give a high appearance salience purely by chance. It tests the hypothesis that the observed call-type count distribution is a sample from the prior distribution.
In step 3400, the salient phone-phrases are clustered into acoustic morphemes by the clustering device 230.
The example below illustrates this acoustic morpheme generation process. Consider a candidate phone-phrase,
f=p1p2 . . . pn,
where pi are phones. Denote its frequency by F(f). A measure of its utility for recognition is the mutual information of its components, denoted MI(f), which may be approximated via
MI(f)=MI(p 1 p 2 . . . p n-1 ;p n).
As discussed above, a simplified measure of its salience for the task is the maximum of the a posteriori distribution,
where C varies over the 15 call-types in the HMIHY task.
These features for phone-phrases observed in the noise-free case are characterized transcr-phone-train. In
For each of these phone-phrases, Pmax(f) is computed, which is a measure of the salience of a phrase for the task.
The goal of this process is to grow the phone-phrases until they have the salience of words and word-phrases. Thus, the search criteria for selecting longer units is a combination of their utility for within-language prediction, as measured by MI, and their utility for the task, as measured by Pmax. For phrases with large Pmax, the MI of the phrase tends to be larger than average. This correlation was exploited successfully for frequency-compensated salience in earlier experiments discussed above, but is not exploited here. In the earlier experiments, a set of salient phone-phrases of length≦16 was generated via a two-pass process as follows:
Select phone-phrases with F(f)≧5 and length≧4;
Filter the training corpus ASR-phone-train with those phrases, using a left-right top-down filter with the phrases prioritized by length.
Select subsequences from the filtered corpus of fragment-length≦4, (i.e. with #phones≦16) and with MI≧1 and Pmax≧0.5.
This particular iterative selection process was selected based on ease of implementation should not be considered optimal. The resultant set of salient phone-phrases have lengths≦16, distributed as shown in
Non-verbal speech may include but are not limited to gestures, body movements, head movements, non-responses, text, keyboard entries, keypad entries, mouse clicks, DTMF codes, pointers, stylus, cable set-top box entries, graphical user interface entries and touchscreen entries, or a combination thereof. Multimodal information is received using multiple channels (i.e., aural, visual, etc.). The user's input communication may also be derived from the verbal and non-verbal speech and the user's or the machine's environment. Basically, any manner of communication falls within the intended scope of the invention. However, for ease of discussion, the invention will be described with respect to verbal speech in the examples and embodiments set forth below.
Working with the output of the non-perfect input speech recognizer 150 introduces the problem of the coverage of the user's input with the acoustic morphemes. To counter this problem, in step 5200, a lattice structure is formed to improve the coverage of the recognizer 150. Lattices are efficient representations of a distribution of alternative hypothesis. A simple example of a lattice network, resulting from the utterance “collect call”, is shown in
In speech recognition, the weights (likelihoods) of the paths of the lattices are interpreted as negative logarithms of the probabilities. For practical purposes, the pruned network will be considered. In this case, the beam search is restricted in the lattice output, by considering only the paths with probabilities above a certain threshold relative to the best path. The threshold r is defined as: ri≦r, with ri=pi/p1, where pi is the probability of the ith path and p1 is the probability of the best path.
In order to quantify the coverage, the number of test sentences are first measured with no detected occurrences of acoustic morphemes, for the experiments using best paths, pruned lattices and full lattices. These numbers are illustrated in
The relative frequency distributions of the number of detected acoustic morphemes in the best paths, in the pruned lattices, and in the full network experiments, are shown in
As shown above, expanding the search of the acoustic morphemes in the lattice network results in improved coverage of the test sentences. It is of course accompanied by an increased number of false detections of the acoustic morphemes. In order to study in more detail the false detection issue, it is beneficial to focus on one particular morpheme Fc, strongly associated to the call-type c=collect. Its occurrences in the best paths will be compared with its occurrences in the lattice network. The salient phone-phrases clustered in this morpheme represent variations of the phrase “collect call”. A subgraph of this morpheme was shown in
In step 5250, the acoustic morpheme detector 160 detects acoustic morphemes that have been recognized and formed into a lattice structure by the input speech recognizer 150 using the acoustic morphemes stored in the acoustic morpheme database 140. In step 5300, the task classification processor 170 performs task classifications based on the detected acoustic morphemes. The task classification processor 170 may apply a confidence function based on the probabilistic relation between the recognized acoustic morphemes and selected task objectives, for example. In step 5400, the task classification processor 170 determines whether a task can be classified based on the detected acoustic morpheme. If the task can be classified, in step 5700, the task classification processor 170 routes the user/customer according to the classified task objective. The process then goes to step 5900 and ends.
If the task cannot be classified in step 5400 (i.e. a low confidence level has been generated), in step 5500, a dialog module (located internally or externally) the task classification processor 170 conducts dialog with the user/customer to obtain clarification of the task objective. After dialog has been conducted with the user/customer, in step 5600, the task classification processor 170 determines whether the task can now be classified based on the additional dialog. If the task can be classified, the process proceeds to step 5700 and the user/customer is routed in accordance with the classified task objective and the process ends at step 5900. However, if task can still not be classified, in step 5800, the user/customer is routed to a human for assistance and then the process goes to step 5900 and ends
An experiment evaluating the utility of these methods in the HMIHY task was conducted. A classifier was trained from the output of a phone recognizer on 7462 utterances, which was denoted ASR-phone-train. Salient phone-phrases of length≦16 were selected, as described above. The salient phone-phrases were then clustered into salient grammar fragments. A single-layer neural net was trained with these fragments as input features. The resultant classifier was applied to the 1000 utterance test-set, ASR-phone-test.
The call-classification results are scored following the methodology of U.S. Pat. No. 5,675,707. In this method, an utterance is classified by the system as one of 14 call-types or rejected as ‘other’. Rejection is based on a salience-threshold for the resulting classification. One dimension of performance is the False Rejection Rate (FRR), which is the probability that an utterance is rejected in the case that the user wanted one of the call-types. The cost of such an error is a lost opportunity for automation. The second dimension of performance is the Probability of Correct Classification (Pc) when the machine attempts a decision. The cost of such an error is that of recovery via dialog. Varying the rejection threshold traces a performance curve with axes Pc and FRR.
Searching in the lattice network will introduce the additional problem of multiple detections of the acoustic morphemes on different levels of the lattice network, and the issue of combining them optimally. For an evaluation of the usefulness of the phone lattices, the problem of treating multiple detections will be deferred from different levels of the network, and stop our search in the lattice network as soon as a sentence with one or more detections is found. The existing call-type classifier is modified in the following way: for the test sentences without detected occurrences of the acoustic morphemes in the best paths, the search is expanded in the lattice network and stops as soon as a sentence with one or more detections is found.
As shown in
While the invention has been described with reference to the above embodiments, it is to be understood that these embodiments are purely exemplary in nature. Thus, the invention is not restricted to the particular forms shown in the foregoing embodiments. Various modifications and alterations can be made thereto without departing from the spirit and scope of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4777600||Jul 28, 1986||Oct 11, 1988||Kabushiki Kaisha Toshiba||Phonetic data-to-kanji character converter with a syntax analyzer to alter priority order of displayed kanji homonyms|
|US4827521 *||Mar 27, 1986||May 2, 1989||International Business Machines Corporation||Training of markov models used in a speech recognition system|
|US4866778||Aug 11, 1986||Sep 12, 1989||Dragon Systems, Inc.||Interactive speech recognition apparatus|
|US4882759 *||Apr 18, 1986||Nov 21, 1989||International Business Machines Corporation||Synthesizing word baseforms used in speech recognition|
|US4903305||Mar 23, 1989||Feb 20, 1990||Dragon Systems, Inc.||Method for representing word models for use in speech recognition|
|US5029214||Aug 11, 1986||Jul 2, 1991||Hollander James F||Electronic speech control apparatus and methods|
|US5033088||Dec 21, 1989||Jul 16, 1991||Voice Processing Corp.||Method and apparatus for effectively receiving voice input to a voice recognition system|
|US5062047||Apr 28, 1989||Oct 29, 1991||Sharp Kabushiki Kaisha||Translation method and apparatus using optical character reader|
|US5099425||Sep 11, 1989||Mar 24, 1992||Matsushita Electric Industrial Co., Ltd.||Method and apparatus for analyzing the semantics and syntax of a sentence or a phrase|
|US5210689||Dec 28, 1990||May 11, 1993||Semantic Compaction Systems||System and method for automatically selecting among a plurality of input modes|
|US5212730||Jul 1, 1991||May 18, 1993||Texas Instruments Incorporated||Voice recognition of proper names using text-derived recognition models|
|US5297039||Jan 27, 1992||Mar 22, 1994||Mitsubishi Denki Kabushiki Kaisha||Text search system for locating on the basis of keyword matching and keyword relationship matching|
|US5323316||Feb 1, 1991||Jun 21, 1994||Wang Laboratories, Inc.||Morphological analyzer|
|US5337232||Mar 18, 1992||Aug 9, 1994||Nec Corporation||Morpheme analysis device|
|US5384892||Dec 31, 1992||Jan 24, 1995||Apple Computer, Inc.||Dynamic language model for speech recognition|
|US5390272||Aug 31, 1993||Feb 14, 1995||Amphenol Corporation||Fiber optic cable connector with strain relief boot|
|US5434906||Sep 13, 1993||Jul 18, 1995||Robinson; Michael J.||Method and apparatus for processing an incoming call in a communication system|
|US5457768||Aug 12, 1992||Oct 10, 1995||Kabushiki Kaisha Toshiba||Speech recognition apparatus using syntactic and semantic analysis|
|US5544050||Aug 31, 1993||Aug 6, 1996||Hitachi, Ltd.||Sign language learning system and method|
|US5619410||Mar 29, 1994||Apr 8, 1997||Nec Corporation||Keyword extraction apparatus for Japanese texts|
|US5651095||Feb 8, 1994||Jul 22, 1997||British Telecommunications Public Limited Company||Speech synthesis using word parser with knowledge base having dictionary of morphemes with binding properties and combining rules to identify input word class|
|US5666400||Jul 7, 1994||Sep 9, 1997||Bell Atlantic Network Services, Inc.||Intelligent recognition|
|US5675707||Sep 15, 1995||Oct 7, 1997||At&T||Automated call router system and method|
|US5719921||Feb 29, 1996||Feb 17, 1998||Nynex Science & Technology||Methods and apparatus for activating telephone services in response to speech|
|US5752230||Aug 20, 1996||May 12, 1998||Ncr Corporation||Method and apparatus for identifying names with a speech recognition program|
|US5794193||Sep 15, 1995||Aug 11, 1998||Lucent Technologies Inc.||Automated phrase generation|
|US5832480||Jul 12, 1996||Nov 3, 1998||International Business Machines Corporation||Using canonical forms to develop a dictionary of names in a text|
|US5839106||Dec 17, 1996||Nov 17, 1998||Apple Computer, Inc.||Large-vocabulary speech recognition using an integrated syntactic and semantic statistical language model|
|US5860063||Jul 11, 1997||Jan 12, 1999||At&T Corp||Automated meaningful phrase clustering|
|US5878390||Jun 23, 1997||Mar 2, 1999||Atr Interpreting Telecommunications Research Laboratories||Speech recognition apparatus equipped with means for removing erroneous candidate of speech recognition|
|US5905774||Mar 10, 1997||May 18, 1999||Stentor Resource Centre, Inc.||Method and system of accessing and operating a voice message system|
|US6006186 *||Oct 16, 1997||Dec 21, 1999||Sony Corporation||Method and apparatus for a parameter sharing speech recognition system|
|US6021384||Oct 29, 1997||Feb 1, 2000||At&T Corp.||Automatic generation of superwords|
|US6023673||Jun 4, 1997||Feb 8, 2000||International Business Machines Corporation||Hierarchical labeler in a speech recognition system|
|US6044337||Oct 29, 1997||Mar 28, 2000||At&T Corp||Selection of superwords based on criteria relevant to both speech recognition and understanding|
|US6138085||Jul 31, 1997||Oct 24, 2000||Microsoft Corporation||Inferring semantic relations|
|US6173261||Dec 21, 1998||Jan 9, 2001||At&T Corp||Grammar fragment acquisition using syntactic and semantic clustering|
|US6192110||Oct 3, 1997||Feb 20, 2001||At&T Corp.||Method and apparatus for generating sematically consistent inputs to a dialog manager|
|US6208964||Aug 31, 1998||Mar 27, 2001||Nortel Networks Limited||Method and apparatus for providing unsupervised adaptation of transcriptions|
|US6233553||Sep 4, 1998||May 15, 2001||Matsushita Electric Industrial Co., Ltd.||Method and system for automatically determining phonetic transcriptions associated with spelled words|
|US6243683||Dec 29, 1998||Jun 5, 2001||Intel Corporation||Video control of speech recognition|
|US6308156||Mar 8, 1997||Oct 23, 2001||G Data Software Gmbh||Microsegment-based speech-synthesis process|
|US6311152||Apr 8, 1999||Oct 30, 2001||Kent Ridge Digital Labs||System for chinese tokenization and named entity recognition|
|US6317707||Dec 7, 1998||Nov 13, 2001||At&T Corp.||Automatic clustering of tokens from a corpus for grammar acquisition|
|US6397179||Nov 4, 1998||May 28, 2002||Nortel Networks Limited||Search optimization system and method for continuous speech recognition|
|US6895377||Mar 23, 2001||May 17, 2005||Eliza Corporation||Phonetic data processing system and method|
|US7085720 *||Oct 18, 2000||Aug 1, 2006||At & T Corp.||Method for task classification using morphemes|
|US20040199375||Apr 5, 2004||Oct 7, 2004||Farzad Ehsani||Phrase-based dialogue modeling with particular application to creating a recognition grammar for a voice-controlled user interface|
|1||A. L. Gorin, "Processing of Semantic Information In Fluently Spoken Language", To appear in the Proceedings of the Intl. Conf. On Spoken Language Processing (CSLP), Oct. 1996, AT&T Research. pp. 1-4.|
|2||A. L. Gorin, et al., "An Experiment In spoken Language Acquisition", IEEE Transactions on Speech and Audio Processing, vol. 2, No. 1, Par II, Jan. 1994, pp. 224-240.|
|3||A. L. Gorin, et al., "Automated Call Routing In A Telecommunications Network", Proceedings, Second IEEE Workshop On Interactive Voice Technology For Telecommunications Applications, Sep. 26-27, 1994, pp. 137-140.|
|4||A. L. Gorin, et al., "On Adaptive Acquisition of Spoken Language", Neural Networks for Signal Processing, 1991; 1991, pp. 422-431.|
|5||A. L. Gorin, et al., "Spoken Dialog As A Feedback Control System", ESCA Workshop on Spoken Dialogue Systems-Theories and Applications, Vigso, Denmark, May 30-Jun. 2, 1995, AT&T Bell Laboratories, pp. 173-175.|
|6||A. L. Gorin, et al., "Spoken Language Acquisition For Automated Call Routing In A Telephone Network", Proceedings 1993 IEEE Workshop on Automatic Speech Recognition, Dec. 12-15, 1993, p. 99-100.|
|7||A. L. Gorin, et al., "Spoken Language Acquisition For Automated Call Routing", ICSLP 94, 1994 International Conference of Spoken Language Processing, Sep. 18-22, 1994, Pacific Convention Plaza Yokohama, pp. 1483-1486.|
|8||A.L. Gorin et al. "How May I Help You?", AT&T Labs Research, p. 1-4, as it was to appear in the Proc. Of IVITA, Philadelphia, Oct. 1996.|
|9||Allen Gorin, "On Automated Language Acquisition" J. Accoust. Soc. Am. 97 (6) Jun. 1995, pp. 3441-3461.|
|10||Irene Langkilde, et al, "Automatic Prediction of Problematic Human-Computer Dialogues in How May I Help You?", AT&T Labs Research, p. 1-4.|
|11||Laura G. Miller, et al., "A Structured Network Architecture For Adaptive Language Acquisition", ICASSP '92: Acoustics, Speech & Signal Processing Conference; 1992, vol. 1, I-201-I204.|
|12||*||Nakamura et al.; A High-speed Morpheme Extraction System Using Dictionary Database; Data Engineering, 1988. Proceedings. Fourth International Conference on Feb. 1-5, 1988 pp. 488-495.|
|13||Ponceleon, et al, "Automatic Discovery of Salient Segments in Imperfect Speech Transcripts", CIKM '01, Atlanta, Georgia, Nov. 5-10, 2001.|
|14||Ron Cole, et al. "The Challenge of Spoken Language Systems: I Research Direction For The Nineties", IEEE Transactions on Speech and Audio Processing; I/95; vol. 3, Issue 1, pp. 1-21.|
|15||U.S. Appl. No. 08/943,944, Method and Apparatus for Generating Semantically Consistent Inputs to a Dialog Manager, filed Oct. 3, 1997, by Alicia Abella, et al.|
|16||U.S. Appl. No. 09/690,721, "Method for Task Classification Using Morphemes", filed Oct. 18, 2001, by Allen L. Gorin, et al.|
|17||U.S. Appl. No. 09/690,903, "Method for Generating Morphemes", filed Oct. 18, 2000, by Allen L. Gorin, et al.|
|18||U.S. Appl. No. 09/699,494, "Method for Automated Task Classification and Routing", filed Oct. 31, 2000, by A. Gorin. et al.|
|19||U.S. Appl. No. 09/699,495, "Automated Task Classification System", filed Oct. 31, 2000, by A. Gorin et al.|
|20||U.S. Appl. No. 09/699,496, "System and Method for Conducting Dialog with a User in an Automatic Task Classification System", filed Oct. 31, 2000, by A. Gorin, et al.|
|21||U.S. Appl. No. 09/712,192, "Method and System for Predicting Problematic Situations in Automated Dialog", filed Nov. 15, 2000, by Gorin, et al.|
|22||U.S. Appl. No. 09/712,194, "Method and System for Predicting Problematic Dialog Situations in a Task Classification System", filed Nov. 15, 2000, by Gorin, et al..|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7440893 *||Sep 6, 2005||Oct 21, 2008||At&T Corp.||Automated dialog method with first and second thresholds for adapted dialog strategy|
|US7487088 *||May 12, 2006||Feb 3, 2009||At&T Intellectual Property Ii, L.P.||Method and system for predicting understanding errors in a task classification system|
|US7529667 *||Sep 6, 2005||May 5, 2009||At&T Intellectual Property Ii||Automated dialog system and method|
|US7873519 *||Oct 31, 2007||Jan 18, 2011||Phoenix Solutions, Inc.||Natural language speech lattice containing semantic variants|
|US7933773 *||Mar 25, 2009||Apr 26, 2011||At&T Intellectual Property Ii, L.P.||Natural language understanding monitoring system for identifying a task|
|US8200491 *||Aug 27, 2011||Jun 12, 2012||At&T Intellectual Property Ii, L.P.||Method and system for automatically detecting morphemes in a task classification system using lattices|
|US8280741 *||Feb 28, 2008||Oct 2, 2012||Airbus Operations Sas||Method and system for the entry of flight data for an aircraft, transmitted between a crew on board the aircraft and ground staff|
|US8301448 *||Mar 29, 2006||Oct 30, 2012||Nuance Communications, Inc.||System and method for applying dynamic contextual grammars and language models to improve automatic speech recognition accuracy|
|US8392188||Sep 21, 2001||Mar 5, 2013||At&T Intellectual Property Ii, L.P.||Method and system for building a phonotactic model for domain independent speech recognition|
|US8612212||Mar 4, 2013||Dec 17, 2013||At&T Intellectual Property Ii, L.P.||Method and system for automatically detecting morphemes in a task classification system using lattices|
|US8909529||Nov 15, 2013||Dec 9, 2014||At&T Intellectual Property Ii, L.P.||Method and system for automatically detecting morphemes in a task classification system using lattices|
|US9002710||Sep 12, 2012||Apr 7, 2015||Nuance Communications, Inc.||System and method for applying dynamic contextual grammars and language models to improve automatic speech recognition accuracy|
|US9280969||Jun 10, 2009||Mar 8, 2016||Microsoft Technology Licensing, Llc||Model training for automatic speech recognition from imperfect transcription data|
|US9311914 *||Sep 3, 2012||Apr 12, 2016||Nice-Systems Ltd||Method and apparatus for enhanced phonetic indexing and search|
|US9502032 *||Oct 28, 2014||Nov 22, 2016||Google Inc.||Dynamically biasing language models|
|US9514126||Nov 13, 2014||Dec 6, 2016||At&T Intellectual Property Ii, L.P.|
|US20050141391 *||Feb 16, 2005||Jun 30, 2005||Tetsuo Ueyama||Optical pickup|
|US20070233488 *||Mar 29, 2006||Oct 4, 2007||Dictaphone Corporation||System and method for applying dynamic contextual grammars and language models to improve automatic speech recognition accuracy|
|US20080221886 *||Feb 28, 2008||Sep 11, 2008||Airbus France||Method and system for the entry of flight data for an aircraft, transmitted between a crew on board the aircraft and ground staff|
|US20090138266 *||Aug 29, 2008||May 28, 2009||Kabushiki Kaisha Toshiba||Apparatus, method, and computer program product for recognizing speech|
|US20100318355 *||Jun 10, 2009||Dec 16, 2010||Microsoft Corporation||Model training for automatic speech recognition from imperfect transcription data|
|US20110313769 *||Aug 27, 2011||Dec 22, 2011||At&T Intellectual Property Ii, L.P.||Method and System for Automatically Detecting Morphemes in a Task Classification System Using Lattices|
|US20140067373 *||Sep 3, 2012||Mar 6, 2014||Nice-Systems Ltd||Method and apparatus for enhanced phonetic indexing and search|
|US20160104482 *||Oct 28, 2014||Apr 14, 2016||Google Inc.||Dynamically biasing language models|
|U.S. Classification||704/254, 704/243|
|Jun 20, 2002||AS||Assignment|
Owner name: AT&T CORP., NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GORIN, ALLEN LOUIS;PETROVSKA-DELACRETAZ, DIJANA;RICCARDI, GIUSEPPE;AND OTHERS;REEL/FRAME:013018/0832;SIGNING DATES FROM 20020521 TO 20020614
|Mar 23, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Mar 25, 2015||FPAY||Fee payment|
Year of fee payment: 8
|Mar 28, 2016||AS||Assignment|
Owner name: AT&T INTELLECTUAL PROPERTY II, L.P., GEORGIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:AT&T PROPERTIES, LLC;REEL/FRAME:038274/0917
Effective date: 20160204
Owner name: AT&T PROPERTIES, LLC, NEVADA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:AT&T CORP.;REEL/FRAME:038274/0841
Effective date: 20160204